Feat/kzg dummy srs

[ThomasPiellard]

Description

When creating a KZG SRS, setting alpha=-1 in NewSRS now generates quickly a balanced, valid SRS. The SRS is in fact <αⁱ[G₁]> where α is of order 4. We cannot use <[G₁]> as the benchmarks are tampered when committing a polynomial (the multi scalar multiplication is twice faster if we do that).

See this issue in gnark.

Type of change

• New feature (non-breaking change which adds functionality)

How has this been tested?

TestVerifySinglePointQuickSRS

How has this been benchmarked?

BenchmarkSRSGen

BenchmarkSRSGen/real_SRS-12         	       2	 549265787 ns/op
BenchmarkSRSGen/quick_SRS-12        	    1540	    787304 ns/op

BenchmarkKZGCommit -> we see that the balanced SRS behaves like a normal one

BenchmarkKZGCommit/real_SRS-12         	              50	  24403057 ns/op
BenchmarkKZGCommit/quick_SRS-12        	      50	  23168466 ns/op

[github-actions]

Summary

✅ Passed: 5923
❌ Failed: 0
🚧 Skipped: 5

🚧 Skipped

• TestReference (github.com/consensys/gnark-crypto/ecc/bn254/fr/sis)
• TestLimbDecomposition (github.com/consensys/gnark-crypto/ecc/bn254/fr/sis)
• TestAppend (github.com/consensys/gnark-crypto/ecc/bn254/fr/tensor-commitment)
• TestAppendSis (github.com/consensys/gnark-crypto/ecc/bn254/fr/tensor-commitment)
• TestCommitmentSis (github.com/consensys/gnark-crypto/ecc/bn254/fr/tensor-commitment)